per hour and this schedule must be adhered to closely in order that it may be possible to check the results in other firings. The specimens must be so placed that they can be reached from without the furnace and withdrawn by means of suitable tongs. It is also well to place a series of pyrometric cones as close to the briquettes as possible to serve as a check on the heat treatment. When the pyrometer indicates the temperature of 1,150°C. the first briquette should be withdrawn and at once placed in an auxiliary muffle furnace maintained at a temperature of about 600 to 700°C. If the temperature interval decided upon is 25°C. the next specimen must be removed at 1,175° and so on until the maximum temperature of 1,425° has been reached. The temperature intervals may be taken as large or small as is convenient. Where but few points are desired they may be made as large as 100°; a convenient interval is 40° which roughly approximates two standard cones. When all of the briquettes have been removed and placed in the auxiliary muffle the latter is allowed to cool down at its natural rate. When cold the specimens are marked according to the temperature at which they have been withdrawn and their volumes determined.
For the purpose of securing comparable records it is customary to compute the percentage of shrinkage of each specimen based upon its volume in the dried state
W DEGRESS CZNTI&RADZ.
I(1iu. 3.—Clay shrinkage on heating.
according to the obvious relation: 100(z>i — vz)/vi, where vi = volume of briquette in the dried state and v2 = its volume in the fired state. The plotting of these results in graphical form oilers a convenient form of comparison from which an estimate of the behavior of the material can be readily made. Such a graph is shown in the upper curve of Fig. 3 which at once tells us that the refractory continues to contract until the temperature 1,350°C. has been reached. The material therefore must be burned to this point if it is desired that it be as constant in volume as possible. It will be no tod that beyond this temperature the shrinkage decreases which is due to expansion of the mass, and the downward trend of the curve is equivalent to negative shrinkage, or expansion. This phenomenon is very characteristic of clays which "overfire," i.e. which on account of excessive softening due to heat and the evolution of gases develop a vesicular or spongy structure accompanied by a swelling of the mass. Clay in this state suffers a decided deterioration in quality since it flows and deforms, possesses low mechanical strength and offers but little resistance to corroding influences such as slags. The diagram thus gives important data concerning a refractory, namely the rate of shrinkage, with temperature, dv/dt, the temperature at which maximum condensation has taken place and the point at which "overfiring" begins.
It is evident that such studies show also volume alterations due to molecular changes, such as the transformation of quartz to cristobalite and tridymite, the decomposition of clay into sillimanitej etc. 31